Maxnet output produced using SDMtune version 1.3.1 (Vignali S. et al., 2020).

ROC curve

## Warning: The following aesthetics were dropped during statistical transformation: m, d
## ℹ This can happen when ggplot fails to infer the correct grouping structure in the data.
## ℹ Did you forget to specify a `group` aesthetic or to convert a numerical variable into a factor?

Thresholds

Threshold Cloglog value Fractional predicted area Training omission rate Test omission rate P-values
Minimum training presence 0.06823105 0.380 0.00000 0.0009920635 0
Equal training sensitivity and specificity 0.37706689 0.138 0.12500 0.0019841270 0
Maximum training sensitivity plus specificity 0.30189623 0.161 0.06250 0.0009920635 0
Equal test sensitivity and specificity 0.35417161 0.143 0.09375 0.0009920635 0
Maximum test sensitivity plus specificity 0.35417161 0.143 0.09375 0.0009920635 0

Distribution map

Variable importance

Variable Permutation_importance sd
Bio2 31.6 0.022
Bio19 21.3 0.027
Bio9 13.7 0.015
Bio3 13.0 0.011
Bio18 11.3 0.019
Bio15 9.0 0.015

Jackknife test

Model settings

Shown below the settings used to train the model:

  • Model type: Maxnet
  • Train locations: 1032
    • presence: 32
    • absence/background: 1000
  • Test locations: 1008
    • presence: 8
    • absence/background: 1000
  • Continuous variables: Bio2, Bio3, Bio9, Bio15, Bio18, Bio19
  • Categorical variables:
  • Output type: cloglog
  • Feature Class combination: lqpht
  • Regularization multiplier: 0.3
  • Do clamping for predictions: TRUE

References

  • Vignali, S., Barras, A. G., Arlettaz, R., Braunisch, V. (2020). SDMtune: An R package to tune and evaluate species distribution models. Ecology and Evolution, 10(20), 11488–11506. https://doi.org/10.1002/ece3.6786
  • Vignali, S., Barras, A. G., Braunisch, V. (2023). SDMtune: Species Distribution Model Selection. R package version 1.3.1. https://CRAN.R-project.org/package=SDMtune